The project is built upon a quantitative sensori-motor learning task in goldfish which is provided with controls for non-learning variables. Training in the task results in selective labelling of discrete brain RNAs which are messenger-type, as judged by size, kinetic, and structural parameters. Training also results in increased phosphorylation of brain nuclear non-histone proteins. Because phosphorylation of this class of proteins is known to correspond with selective activation of the genome in other systems, the changes in brain protein after training are consistent with the RNA changes. Work under way attempts to define the relation of these molecular changes to the learning component of the task, and to strengthen the definition by providing appropriate pharmacological-behavioral correlates. The objective of the work is understanding the basic molecular mechanisms of brain function. BIBLIOGRAPHIC REFERENCES: Kaplan and Sirlin, Training induced alteration in leucine transfer RNA of goldfish brain, Brain Research 83, 451, 1975. Sirlin, Time-dependent differences in the brain RNA population of goldfish after behaviroal training, Soc. Neurosci. Fifth Ann. Meeting 332, 1975 (abstract).